The Straw Poll that Broke the Camel's Back

As a process consultant I get the chance to observe first hand which methods consensus groups use to work their way through issues. One of the most common is the straw poll, employed to determine which way and how strongly the wind is blowing part way through a discussion. As a consensus trainer I cordially detest straw polls, and today I want to make the case for why this is not a good practice.

[Years ago JRR Tolkien wrote that he "cordially detested" allegory when responding to a suggestion that Lord of the Rings was written with Hitler as the prototype for Sauron, and I've been nurturing that turn of phrase ever since, hoping that I'd eventually be able to dust it off and put it back into play. Today I finally found the right occasion.]

Think of me as the Big Bad Process Lupine who is going to huff and puff and attempt to blow down the house of straw... (OK, so I get carried away with the metaphors.)

Consensus is a process that is all together different from Voting. While Consensus is based on the concept that the best decisions will emerge from the full group being in alignment about how to proceed, Voting is based on the idea that the best proposal will emerge from a healthy competition.

In Consensus a proposal does not advance to acceptance in the presence of any principled objections—even one; in Voting it only takes a majority of votes for a proposal to succeed. Living in the US, nearly everyone has experience with parliamentary procedure and democratic decision-making that relies on majority rule. While there are number of possible variations, in the main Voting works like this: proposals are put forward, their merits are debated, and eventually there's a vote. If one proposal garners a majority, it passes and the matter is settled.

One of the reasons I'm uncomfortable with consensus groups using straw polls is that it's a form of voting (albeit a non-binding one), and one of the more common difficulties that groups have in fully realizing the potential of Consensus is that they never succeed in creating a culture of collaboration (perhaps because they don't even perceive the need for it). If a group attempts to superimpose Consensus on a culture of Voting, then you're just talking about unanimous voting, and it's no wonder that many groups report frustration and weak results (as the only proposals that can jump that high bar are ones are typically so watered down as to have little potency for addressing issues).

Thus, I'm highly concerned that if a consensus group uses straw polls, they'll be keeping alive a competitive dynamic that undercuts the attempt to build and maintain the requisite collaborative culture.

The point of straw polls is to test for the presence of momentum favoring one response to an issue over another. The idea is that this will clear the fog and help the group move productively through the forest (or at least the thicket) of ideas. While there is undoubtedly a need for groups to know where they are in a conversation and what aspects hold the most promise for being a path through the woods, I think there are better ways to meet that need than with straw polls.

When a group votes, the intention is that the group will be influenced to move in the direction of the majority. (I know that's not always what happens, but that's what the people who propose straw polls are hoping will happen.) No matter how many times you insist that the straw poll is not binding and is informational only, whenever you vote you are invoking the culture of Voting, and the group can hardly help but be influenced by that dynamic. Those in the majority start to relax (after all, they're winning); those in the minority start to feel the pressure (c'mon, you're holding up progress). Some people who suspect they are in the minority may even alter their voting so as not to be singled out for this kind of attention. To the extent that the group slides back into the culture of Voting, it moves out of the collaborative environment where everyone is working purposefully and trustingly toward a we're-all-on-the-same-team solution that everyone can support.

Better, I think, is for the group (led by the facilitator) to learn to follow the energy of a discussion, diligently identifying and working all relevant ends of the discussion (not just trying to find the road where most of the traffic is). Instead of asking the group which views seem to be dominant (the point of a straw poll), you can ask instead, "What ideas to people have that will bridge the disparate concerns expressed?" In a Voting culture the conversation pivots around advocacy (of one's own position) and challenges (of differing viewpoints). In a Consensus culture, the conversation should revolve around how to develop agreements that balance and connect all the factors. You are looking for how to draw an elegant circle around all the input—not just most of it.

Take some device from the Big Bad Wolf: if want to built an enduring collection of consensus agreements, go light on the straw. It may come at a price that's too heavy.